fedora core 6 1.2949 + vserver 2.2.0
[linux-2.6.git] / fs / ocfs2 / cluster / nodemanager.c
index e1fceb8..b17333a 100644 (file)
@@ -35,7 +35,7 @@
 /* for now we operate under the assertion that there can be only one
  * cluster active at a time.  Changing this will require trickling
  * cluster references throughout where nodes are looked up */
-static struct o2nm_cluster *o2nm_single_cluster = NULL;
+struct o2nm_cluster *o2nm_single_cluster = NULL;
 
 #define OCFS2_MAX_HB_CTL_PATH 256
 static char ocfs2_hb_ctl_path[OCFS2_MAX_HB_CTL_PATH] = "/sbin/ocfs2_hb_ctl";
@@ -97,17 +97,6 @@ const char *o2nm_get_hb_ctl_path(void)
 }
 EXPORT_SYMBOL_GPL(o2nm_get_hb_ctl_path);
 
-struct o2nm_cluster {
-       struct config_group     cl_group;
-       unsigned                cl_has_local:1;
-       u8                      cl_local_node;
-       rwlock_t                cl_nodes_lock;
-       struct o2nm_node        *cl_nodes[O2NM_MAX_NODES];
-       struct rb_root          cl_node_ip_tree;
-       /* this bitmap is part of a hack for disk bitmap.. will go eventually. - zab */
-       unsigned long   cl_nodes_bitmap[BITS_TO_LONGS(O2NM_MAX_NODES)];
-};
-
 struct o2nm_node *o2nm_get_node_by_num(u8 node_num)
 {
        struct o2nm_node *node = NULL;
@@ -152,14 +141,16 @@ static struct o2nm_node *o2nm_node_ip_tree_lookup(struct o2nm_cluster *cluster,
        struct o2nm_node *node, *ret = NULL;
 
        while (*p) {
+               int cmp;
+
                parent = *p;
                node = rb_entry(parent, struct o2nm_node, nd_ip_node);
 
-               if (memcmp(&ip_needle, &node->nd_ipv4_address,
-                          sizeof(ip_needle)) < 0)
+               cmp = memcmp(&ip_needle, &node->nd_ipv4_address,
+                               sizeof(ip_needle));
+               if (cmp < 0)
                        p = &(*p)->rb_left;
-               else if (memcmp(&ip_needle, &node->nd_ipv4_address,
-                               sizeof(ip_needle)) > 0)
+               else if (cmp > 0)
                        p = &(*p)->rb_right;
                else {
                        ret = node;
@@ -541,6 +532,179 @@ static struct o2nm_node_group *to_o2nm_node_group(struct config_group *group)
 }
 #endif
 
+struct o2nm_cluster_attribute {
+       struct configfs_attribute attr;
+       ssize_t (*show)(struct o2nm_cluster *, char *);
+       ssize_t (*store)(struct o2nm_cluster *, const char *, size_t);
+};
+
+static ssize_t o2nm_cluster_attr_write(const char *page, ssize_t count,
+                                       unsigned int *val)
+{
+       unsigned long tmp;
+       char *p = (char *)page;
+
+       tmp = simple_strtoul(p, &p, 0);
+       if (!p || (*p && (*p != '\n')))
+               return -EINVAL;
+
+       if (tmp == 0)
+               return -EINVAL;
+       if (tmp >= (u32)-1)
+               return -ERANGE;
+
+       *val = tmp;
+
+       return count;
+}
+
+static ssize_t o2nm_cluster_attr_idle_timeout_ms_read(
+       struct o2nm_cluster *cluster, char *page)
+{
+       return sprintf(page, "%u\n", cluster->cl_idle_timeout_ms);
+}
+
+static ssize_t o2nm_cluster_attr_idle_timeout_ms_write(
+       struct o2nm_cluster *cluster, const char *page, size_t count)
+{
+       ssize_t ret;
+       unsigned int val;
+
+       ret =  o2nm_cluster_attr_write(page, count, &val);
+
+       if (ret > 0) {
+               if (cluster->cl_idle_timeout_ms != val
+                       && o2net_num_connected_peers()) {
+                       mlog(ML_NOTICE,
+                            "o2net: cannot change idle timeout after "
+                            "the first peer has agreed to it."
+                            "  %d connected peers\n",
+                            o2net_num_connected_peers());
+                       ret = -EINVAL;
+               } else if (val <= cluster->cl_keepalive_delay_ms) {
+                       mlog(ML_NOTICE, "o2net: idle timeout must be larger "
+                            "than keepalive delay\n");
+                       ret = -EINVAL;
+               } else {
+                       cluster->cl_idle_timeout_ms = val;
+               }
+       }
+
+       return ret;
+}
+
+static ssize_t o2nm_cluster_attr_keepalive_delay_ms_read(
+       struct o2nm_cluster *cluster, char *page)
+{
+       return sprintf(page, "%u\n", cluster->cl_keepalive_delay_ms);
+}
+
+static ssize_t o2nm_cluster_attr_keepalive_delay_ms_write(
+       struct o2nm_cluster *cluster, const char *page, size_t count)
+{
+       ssize_t ret;
+       unsigned int val;
+
+       ret =  o2nm_cluster_attr_write(page, count, &val);
+
+       if (ret > 0) {
+               if (cluster->cl_keepalive_delay_ms != val
+                   && o2net_num_connected_peers()) {
+                       mlog(ML_NOTICE,
+                            "o2net: cannot change keepalive delay after"
+                            " the first peer has agreed to it."
+                            "  %d connected peers\n",
+                            o2net_num_connected_peers());
+                       ret = -EINVAL;
+               } else if (val >= cluster->cl_idle_timeout_ms) {
+                       mlog(ML_NOTICE, "o2net: keepalive delay must be "
+                            "smaller than idle timeout\n");
+                       ret = -EINVAL;
+               } else {
+                       cluster->cl_keepalive_delay_ms = val;
+               }
+       }
+
+       return ret;
+}
+
+static ssize_t o2nm_cluster_attr_reconnect_delay_ms_read(
+       struct o2nm_cluster *cluster, char *page)
+{
+       return sprintf(page, "%u\n", cluster->cl_reconnect_delay_ms);
+}
+
+static ssize_t o2nm_cluster_attr_reconnect_delay_ms_write(
+       struct o2nm_cluster *cluster, const char *page, size_t count)
+{
+       return o2nm_cluster_attr_write(page, count,
+                                      &cluster->cl_reconnect_delay_ms);
+}
+static struct o2nm_cluster_attribute o2nm_cluster_attr_idle_timeout_ms = {
+       .attr   = { .ca_owner = THIS_MODULE,
+                   .ca_name = "idle_timeout_ms",
+                   .ca_mode = S_IRUGO | S_IWUSR },
+       .show   = o2nm_cluster_attr_idle_timeout_ms_read,
+       .store  = o2nm_cluster_attr_idle_timeout_ms_write,
+};
+
+static struct o2nm_cluster_attribute o2nm_cluster_attr_keepalive_delay_ms = {
+       .attr   = { .ca_owner = THIS_MODULE,
+                   .ca_name = "keepalive_delay_ms",
+                   .ca_mode = S_IRUGO | S_IWUSR },
+       .show   = o2nm_cluster_attr_keepalive_delay_ms_read,
+       .store  = o2nm_cluster_attr_keepalive_delay_ms_write,
+};
+
+static struct o2nm_cluster_attribute o2nm_cluster_attr_reconnect_delay_ms = {
+       .attr   = { .ca_owner = THIS_MODULE,
+                   .ca_name = "reconnect_delay_ms",
+                   .ca_mode = S_IRUGO | S_IWUSR },
+       .show   = o2nm_cluster_attr_reconnect_delay_ms_read,
+       .store  = o2nm_cluster_attr_reconnect_delay_ms_write,
+};
+
+static struct configfs_attribute *o2nm_cluster_attrs[] = {
+       &o2nm_cluster_attr_idle_timeout_ms.attr,
+       &o2nm_cluster_attr_keepalive_delay_ms.attr,
+       &o2nm_cluster_attr_reconnect_delay_ms.attr,
+       NULL,
+};
+static ssize_t o2nm_cluster_show(struct config_item *item,
+                                 struct configfs_attribute *attr,
+                                 char *page)
+{
+       struct o2nm_cluster *cluster = to_o2nm_cluster(item);
+       struct o2nm_cluster_attribute *o2nm_cluster_attr =
+               container_of(attr, struct o2nm_cluster_attribute, attr);
+       ssize_t ret = 0;
+
+       if (o2nm_cluster_attr->show)
+               ret = o2nm_cluster_attr->show(cluster, page);
+       return ret;
+}
+
+static ssize_t o2nm_cluster_store(struct config_item *item,
+                                  struct configfs_attribute *attr,
+                                  const char *page, size_t count)
+{
+       struct o2nm_cluster *cluster = to_o2nm_cluster(item);
+       struct o2nm_cluster_attribute *o2nm_cluster_attr =
+               container_of(attr, struct o2nm_cluster_attribute, attr);
+       ssize_t ret;
+
+       if (o2nm_cluster_attr->store == NULL) {
+               ret = -EINVAL;
+               goto out;
+       }
+
+       ret = o2nm_cluster_attr->store(cluster, page, count);
+       if (ret < count)
+               goto out;
+out:
+       return ret;
+}
+
 static struct config_item *o2nm_node_group_make_item(struct config_group *group,
                                                     const char *name)
 {
@@ -550,7 +714,7 @@ static struct config_item *o2nm_node_group_make_item(struct config_group *group,
        if (strlen(name) > O2NM_MAX_NAME_LEN)
                goto out; /* ENAMETOOLONG */
 
-       node = kcalloc(1, sizeof(struct o2nm_node), GFP_KERNEL);
+       node = kzalloc(sizeof(struct o2nm_node), GFP_KERNEL);
        if (node == NULL)
                goto out; /* ENOMEM */
 
@@ -622,10 +786,13 @@ static void o2nm_cluster_release(struct config_item *item)
 
 static struct configfs_item_operations o2nm_cluster_item_ops = {
        .release        = o2nm_cluster_release,
+       .show_attribute         = o2nm_cluster_show,
+       .store_attribute        = o2nm_cluster_store,
 };
 
 static struct config_item_type o2nm_cluster_type = {
        .ct_item_ops    = &o2nm_cluster_item_ops,
+       .ct_attrs       = o2nm_cluster_attrs,
        .ct_owner       = THIS_MODULE,
 };
 
@@ -658,8 +825,8 @@ static struct config_group *o2nm_cluster_group_make_group(struct config_group *g
        if (o2nm_single_cluster)
                goto out; /* ENOSPC */
 
-       cluster = kcalloc(1, sizeof(struct o2nm_cluster), GFP_KERNEL);
-       ns = kcalloc(1, sizeof(struct o2nm_node_group), GFP_KERNEL);
+       cluster = kzalloc(sizeof(struct o2nm_cluster), GFP_KERNEL);
+       ns = kzalloc(sizeof(struct o2nm_node_group), GFP_KERNEL);
        defs = kcalloc(3, sizeof(struct config_group *), GFP_KERNEL);
        o2hb_group = o2hb_alloc_hb_set();
        if (cluster == NULL || ns == NULL || o2hb_group == NULL || defs == NULL)
@@ -676,6 +843,9 @@ static struct config_group *o2nm_cluster_group_make_group(struct config_group *g
        cluster->cl_group.default_groups[2] = NULL;
        rwlock_init(&cluster->cl_nodes_lock);
        cluster->cl_node_ip_tree = RB_ROOT;
+       cluster->cl_reconnect_delay_ms = O2NET_RECONNECT_DELAY_MS_DEFAULT;
+       cluster->cl_idle_timeout_ms    = O2NET_IDLE_TIMEOUT_MS_DEFAULT;
+       cluster->cl_keepalive_delay_ms = O2NET_KEEPALIVE_DELAY_MS_DEFAULT;
 
        ret = &cluster->cl_group;
        o2nm_single_cluster = cluster;